Sound recognition and object detection


Could I implement both sound recognition and object detection at the same time + rtsp streaming ?


Hi @AI_Test , this could be done once we have implemented loading the NN model using the SD card feature in our SDK, which will be using RAM instead of Flash.

We will inform you once this feature is up and you will be able to do both Audio Classification and Object Detection at the same time.

Hi. I am just checking if this feature of audio recognition and object detection at the same time is available yet. For example to detect a car with the engine running and RTSP? Object detection works superbly. Thanks.

You can refer to the latest comment for the code AudioNN + ObjDet · Issue #171 · ambiot/ambpro2_arduino · GitHub.

However, this example will not be part of the Arduino SDK, as by default, models are loaded using flash. There may be some confusion if we add this example to SDK. But you can take the code as a reference for your application.

In order to do Audio Classification + Object Detection, you need to load the model using SD card instead of Flash.

Guide: How to apply your AI model – Realtek IoT/Wi-Fi MCU Solutions